Circulating FGF-23 is regulated by 1alpha,25-dihydroxyvitamin D3 and phosphorus in vivo.

Fibroblast growth factor-23 (FGF-23), a novel phosphate-regulating factor, was elevated in hypophosphatemic patients with X-linked hypophosphatemic rickets/osteomalacia and also in patients with chronic kidney disease. These observations suggested the pathophysiological importance of FGF-23 on phosphate homeostasis. However, regulation of FGF-23 production is still unclear. We investigated effects of both dietary phosphorus and 1alpha,25-dihydroxyvitamin D(3) (1alpha,25(OH)(2)D(3)) on circulating FGF-23 in vivo Administration of. 1alpha,25(OH)(2)D(3) dose-dependently increased serum FGF-23 in thyroparathyroidectomized rats without correlating with serum inorganic phosphorus or serum parathyroid hormone. On the other hand, vitamin D receptor null mice had very low serum FGF-23 and did not respond to the 1alpha,25(OH)(2)D(3) administration. These observations suggested 1alpha,25(OH)(2)D(3) directly or indirectly regulates circulating FGF-23. Serum FGF-23 had a strong correlation with serum inorganic phosphorus controlled by dietary phosphorus in 5/6 nephrectomized rats. High phosphate diet elicited a 5-fold increase in serum FGF-23 compared with sham-operated rats, whereas serum FGF-23 did not correlate with serum calcium or serum creatinine in 5/6 nephrectomized rats. Administration of 1alpha,25-dihydroxyvitamin D(3) also elicited a severalfold increase in serum FGF-23 in the uremic rats. Taken together, this shows that both serum phosphorus and 1alpha,25(OH)(2)D(3) regulate circulating FGF-23 independent of each other. Therefore, we proposed there was a feedback loop existing among serum phosphorus, 1alpha,25(OH)(2)D(3), and FGF-23, in which the novel phosphate-regulating bone-kidney axis integrated with the parathyroid hormone-vitamin D(3) axis in regulating phosphate homeostasis.

The purpose of this study was to evaluate the effects of dietary phosphorus and 1␣,25(OH) 2 D 3 on FGF-23 production. Administration of FGF-23 protein or overexpression of Fgf23 gene in rodent suppressed 1␣,25(OH) 2 D 3 production by reducing 25-hydroxyvitamin D 3 1␣-hydroxylase in the proximal tubules (12,(21)(22)(23). On the contrary, Fgf23-null mice reported increased circulating 1␣,25(OH) 2 D 3 despite hyperphosphatemia, hypercalcemia, and low PTH levels (24). Administration of 1␣,25(OH) 2 D 3 increased serum FGF-23 in normal mice (25). These observations suggested mutual regulation between FGF-23 and 1␣,25(OH) 2 D 3 ; however, 1␣,25(OH) 2 D 3 administration also increases intestinal phosphate uptake and suppresses PTH. Thus, we used thyroparathyroidectomized rats as well as 5/6 nephrectomized rats fed a diet with various kinds of phosphorus content to examine the direct effect of 1␣,25(OH) 2 D 3 administration on serum FGF-23. * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Dietary phosphate deprivation or loading rapidly induces activation or repression of phosphate absorption in kidney and in intestine, mainly by inducing or suppressing type II sodium-dependent phosphate (Na/P i ) cotransporter expression (26,27). Therefore, the serum phosphorus level is not susceptible to the change in dietary phosphorus content in normal animals in vivo. We investigated the effects of dietary phosphorus on FGF-23 production in vivo using 5/6 nephrectomized uremic rats, in which the serum phosphorus level can be easily manipulated by dietary phosphorus due to reduced kidney function. We also examined the correlations between serum FGF-23 and (a) serum inorganic phosphorus, (b) serum calcium, (c) serum creatinine, and (d) serum PTH in 5/6 nephrectomized rats.

Thyroparathyroidectomized (TPTX) Rats and PTH-infused TPTX Rats-Eight-week-old male Sprague-Dawley rats purchased from
Charles River (Tokyo, Japan) were TPTX under ether anesthesia. After confirming hypocalcemia had been induced, rats were divided to two groups, the TPTX group and the TPTX ϩ PTH group. Rats in the TPTX ϩ PTH group were subcutaneously implanted with ALZET® osmotic pumps (model 2ML2, Durect Corp., Cupertino, CA) and administered human PTH-(1-34) at a constant rate of 2.4 g/day for 14 days. Rats in the TPTX group were intravenously injected with vehicle, 50 ng/kg or 300 ng/kg 1␣,25(OH) 2 D 3 , three times a week for 2 weeks; rats in the TPTX ϩ PTH group were injected with either vehicle or 50 ng/kg 1␣,25(OH) 2 D 3 three times a week for 2 weeks. Blood samples were collected and various parameters analyzed as described above.
Vitamin D Receptor Null(Ϫ/Ϫ) (VDRKO) Mice-VDRKO mice and their littermates kept on a high calcium and high phosphorus diet (28) were administered vehicle or 300 ng/kg 1␣,25(OH) 2 D 3 three times a week for 2 weeks. Blood samples were collected and analyzed as described above.

5/6 Subtotally Nephrectomized Uremic Rat Models and Diets-Male
Sprague-Dawley rats weighing 180 -200 g were purchased from CREA Japan and maintained under specific pathogen-free conditions with a 12-h light/dark cycle. After acclimating for 1 week, the rats were 5/6 nephrectomized and then allowed unlimited access to normal rodent chow (CE-2, CREA Japan Inc.) and tap water. The phosphate (P i )controlled diets and 1␣,25(OH) 2 D 3 injection started when rat serum creatinine reached within the range of 1.1-1.5 mg/dl. Three P i -controlled diets (Oriental Yeast Co., Ltd., Osaka, Japan) were used in the present study: a high P i diet, containing 0.9% phosphorus, 0.6% calcium; a midrange P i diet, containing 0.6% phosphorus, 0.6% calcium; and a low P i diet, containing 0.2% phosphorus, 0.6% calcium (26,27).
The following groups of rats were studied: 1) 5/6 nephrectomized rats fed either the high, midrange, or low P i diet for 4 weeks, 2) 5/6 nephrectomized rats fed the high, midrange, or low P i diet and injected with 50 ng of 1␣,25(OH) 2 D 3 /kg of bodyweight intravenously twice weekly for 4 weeks, and 3) age-matched sham-operated rats fed normal rat chow (CE-2, CREA Japan) used as a control. Blood samples were obtained from the vena cava under ether anesthesia at day 28 and various parameters analyzed as described above. Serum PTH was determined using a Rat Intact PTH ELISA kit (Immutopics, Inc., San Clemente, CA).
All animal procedures were conducted in accordance with Chugai Pharmaceutical's ethical guidelines for animal care, and all experimental protocols were approved by the Animal Care Committee of the institution.
Statistical Analysis-Data were expressed as means Ϯ S.E., and statistical significance was determined using Student's t test or Dunnett's t test (SAS Preclinical Package, Version 5.0, SAS Institute Japan, Tokyo) unless otherwise indicated. A p value of Ͻ0.05 was considered statistically significant.
Larsson et al. (18) reported phosphate deprivation and/or phosphate loading to normal subjects did not affect serum FGF-23; however, serum phosphorus weakly correlated with serum FGF-23 in predialysis patients with chronic kidney disease. Recent studies also revealed that serum FGF-23 was elevated in patients with end-stage renal disease (16,20,28). In the present study, we investigated the effect of dietary phosphorus on FGF-23 production using 5/6 nephrectomized rats fed the diets with various kinds of phosphorus content. Serum FGF-23 was elevated in uremic rats; however, serum FGF-23 did not clearly correlate with serum creatinine in those rats as was observed in human subjects. Serum phosphorus was well controlled by the dietary phosphorus in 5/6 nephrectomized rats (Fig. 3A). Serum FGF-23 positively correlated with serum phosphorus in those rats (Fig. 3C). In the physiological condition, a high serum phosphorus suppresses 1␣,25(OH) 2 D 3 production in kidney. Thus, the elevation of serum FGF-23 induced by a high P i diet was independent of serum 1␣,25(OH) 2 D 3 . Moreover, serum FGF-23 was drastically elevated by 1␣,25(OH) 2 D 3 administration in 5/6 nephrectomized rats fed with various P i -controlled diets (Fig. 5). However, serum FGF-23 did not correlate with serum calcium, serum creatinine, or serum PTH in those rats (Fig. 6, B-D). These observations suggested that FGF-23 production was mainly regulated by serum phosphorus and serum 1␣,25(OH) 2 D 3 .
Recent studies (16,17) reported that FGF-23 was elevated in some patients with XLH. Serum phosphorus concentrations were negatively correlated with circulating FGF-23 levels in patients with XLH. Moreover, FGF-23 mRNA expression was FIG. 4. Correlations between serum FGF-23 and serum inorganic phosphorus (A), serum calcium (B), and serum creatinine (C) in 5/6 nephrectomized rats fed with P i -controlled diet. 5/6 nephrectomized rats were fed either high P i , midrange P i , or low P i diet for 4 weeks. Serum creatinine, serum inorganic phosphorus, and serum FGF-23 were determined as described under "Experimental Procedures." Open circles represent the data from individual rats.  5. 1␣,25-Dihydroxyvitamin D 3 administration increased circulating FGF-23 in 5/6 nephrectomized rats. 5/6 nephrectomized rats fed P i -controlled diets were given either vehicle or 50 ng/kg 1␣,25-dihydroxyvitamin D 3 intravenously, twice a week for 4 weeks. Serum FGF-23 was determined by ELISA. 1␣,25-Dihydroxyvitamin D 3 injection to 5/6 nephrectomized rats induced a drastic increase in circulating FGF-23 among all diet groups. Each column represents mean Ϯ S.E. (n ϭ 10 rats/group). *, statistically significant difference between vehicle and 1␣,25(OH) 2 D 3 treated in each diet, p Ͻ 0.05 by Student's t test. HPD, high P i diet; MPD, midrange P i diet; LPD, low P i diet.
1␣,  Up-regulates FGF-23 in Vivo enhanced in the calvarial and mandible bones of Hyp-mouse, which is a homologue of human XLH (13). Mutations in PHEX, a phosphate-regulating gene with homology to endopeptidase on the X-chromosome, are responsible for XLH. PHEX mRNA is predominantly expressed in bone and teeth. 1␣,25(OH) 2 D 3 decreased PHEX mRNA and PHEX protein in primary osteoblasts derived from newborn mouse calvaria as well as MC3T3-E1 cells, a mouse osteoblastic cell line, in vitro (29). In addition, PHEX mRNA expression in tibial bone was suppressed by 1␣,25(OH) 2 D 3 administration in 5/6 nephrectomized rats in vivo (30). It is plausible that administration of 1␣,25(OH) 2 D 3 up-regulated circulating FGF-23 levels in 5/6 nephrectomized rats at least partly by down-regulation of PHEX expression in bones.
FGF-23 induces hypophosphatemia by inhibiting both renal and intestinal P i absorption by suppressing NaP i -IIa and -IIb production (3,12,21,22,25). FGF-23 also inhibits 1␣,25(OH) 2 D 3 production in renal proximal tubules, which results in the reduction of intestinal P i absorption and PTH secretion. On the contrary, 1␣,25(OH) 2 D 3 induced an increase in circulating FGF-23, and also loss of vitamin D signaling in VDRKO mice led to very low serum FGF-23. In 5/6 nephrectomized rats, serum phosphorus controlled by dietary phosphorus content positively correlated with serum FGF-23, suggesting an increase in serum phosphorus induces FGF-23 production. We propose that a feedback loop exists between serum phosphorus, 1␣,25(OH) 2 D 3 , and FGF-23, in which the novel phosphate-regulating bone-kidney axis would be integrated with the parathyroid hormone-vitamin D 3 axis in regulating phosphate homeostasis.